“If you really want to follow our work closely, you can read through our mailing list archives which Sun’s Jon Bosak was kind enough to set up for us.”

just in case, i would like to ask you if you know if Jon has any plan to keep updating this archive ( last mail batch was from october/2007 )…

i found very interesting the V1 deliberations and discussions during the pre-vote phase, and may be now we can keep following the discussions of V1 regarding the ECMA(microsoft) 2263 pages (kind of) responses

thank you very much

franco merletti

Just forgot to note that I’ve sent letters on the matter “ODF vs. OOXML” to three more Romanian institutions – just to draw attention:

- The Romanian Association for Standardization, ASRO, the national standards body http://www.asro.ro

- The Romanian Ministry of Information Technology and Communication http://www.mcti.ro

Given the position Microsoft has in Romania, I don’t hold many naive ilussions about what Romania’s official position will be at ISO ballot – but this is useful in drawing attention from the Romanian users…

Now I learn, from your blog, that ASRO is questioning its official members (busineses) about what position
Romania should have at ISO in the ODF matter.

Regards,
Răzvan

I’m Răzvan from Bucharest, Romania.

The explanation for that Microsoft Romania guy, Zoli, “complimenting” you is simple… :

In the last few months I did some efforts to draw attention for the Romanian public about the ODF vs. OOXML matter.

You must know that Romanian average users are accustomised with MS Office for many years, especially because, since 1990, one may buy a pirated copy for $3, on the street. This is somehow convenient for Microsoft, because now piracy rate is starting to lower, but non-IT users are so habituated with Microsoft products that they don’t even imagine an alternative.

Microsoft Romania is keeping this „wall of silence” up – despite OpenOffice.org having many users among IT professionals, you can hardly see information about ODF in the official literature.

So I tried to bring a modest contribution in breaking this silence:

- I put up a blog in Romanian, presenting the matter – please see http://rsandu.myblog.ro. It shows as No. 1 on Google if you use “OpenDocument Romania” as searching criteria.

- I’ve sent some messages on this matter to TIC-Lobby, which is a mailing list where all „official” matters in Romanian IT field are discussed (http://beta.agora.ro/agora-bin/lwgate/TIC-LOBBY/)
Among others, it seems all Microsoft Romania employees, excluding the doorman, are members of this list (since it deals with Romanian Goverment’s investition plans in IT) ;-)

Zoli’s text you’ve pointed to is mainly the response of my first tentatives of breaking the silence.

Regards,
Răzvan

Of course, whether he claims it or not, that’s absurd. I’m one of the anonymous masses who has to put up with Word and is sick of all the single-vendor lock-in. In fact, I was the one who wrote and submitted the infamous Slashdot story featuring this blog when all those formatLikeAncientWordProcessor flags first drew widespread criticism. And there’s no one paying me to do this. I do it because the only thing I hate more than bad code and WTF-specifications is knowing that I can’t fix them.

I have no idea how the example snippets were created. We’ve certainly all seen programming books with syntax errors in the code. These typically are caused by late editing changes, faulty integration scripts or hasty preparation. But if we’re seeing hundreds of similar errors then I’d suspect a search & replace editing error or similar global operation.

In any case these are the kinds of things that should have been caught much earlier in the process, during the Ecma review phase. That is the problem with trying to standardize 6,000 pages in less than one year. It doesn’t matter how smart you are or how much you really want to do it quickly, the results are going to be unacceptable. IMHO, it is a disservice to the talented people involved in the creation of OOXML to have put them in the situation where the forced pace of their review reflects so poorly on their efforts.

-Rob

As Christopher notes, many of the XML samples in the OOXML standard are not valid XML. I know firsthand that is is very easy to make this sort of mistake in hand generated XML. But with OOXML, the authors should be able to unzip Office documents and cut and paste. So where do the samples come from?

From work with HTML, I know that there are always degrees of conformance. With HTML, there are accepted tests to determine strict compliance. Do the RelaxNG documents provided in the OOXML specification fill this role? Can I take something like Schematron and and handful of Office 2007 documents and validate?

My worry is this: Does Microsoft Office 2007 currently generate valid OOXML, or does it generate a superset of OOXML? I don’t expect perfection, but the defect density in their documents had better be approaching ’six sigma’ and improving. Am I crazy/paranoid for even suggesting that Microsoft has already extended a protocol that they are asking us to embrace?

What I did find interesting was Inigo Surguy’s analysis of the code samples in the OOXML spec. He found that, in the WordProcessingML section alone, around 300 of the XML examples do not validate.

That is, over 10% of the examples in one part of the spec are plain wrong. How these sorts (and volume) of errors can be put forward for an international standard is crazy.

Thanks for the translation. I’m flattered that they hold my opinion is such regard as to warrant expenditure of their limited wit to comment on it. But I don’t see why “attempting to influence” is denigrated. This 5-month review of OOXML is part of the standards development and approval process. Unless this review period is to be merely ceremonial, now is the time to review the specification for problems, to document those problems, and to persuade others that these problems need to be fixed. That is what a consensus process entails. For some reason Microsoft seems afraid of this.

-Rob

The first part of the Zoli’s blog says something like:

If you will read Bob Sutor or Rob Weir you’ll understand why I say that. For a company that sustains openness it is curious the impressive effort to denigrate a standard (ECMA Office Open XML). As Stephen McGibbon says, you need a big dose of hypocrisy to do that. Especially when the guys do big efforts to influence the standardization organisms. The procedure is simple: without lying (saying plain lies), those who try such influences link to the texts of above-mentioned or (if they are more evolved) process ideas from there. Bullshit! You’ll say I’m paranoiac. I’ve thought of that myself… but, from a week, I see a sustained activity in TIC-Lobby mailing list related to the same subject: “OpenXML it is bad (it is a standard born from Microsoft specifications) and it must not be standardized by ISO because there is already a standard named ODF which is enough” .
He continues with citations from his previous emails on the mailing list and concludes:
I’ve had enough already! But wait a second. Now I see why this has swelled again: ISO starts to ask for the votes of members organizations and ASRO, as member of ISO collects its members opinions. So this is it!

Bullshit, brothers. Bullshit!

If you need/wish I could translate the missing part but it pretty much the same rhetoric you see in Microsoft speeches, nothing new (to me at least).