Thank you for your effort, and keep up the interesting and illuminating writing.

Yours,

JustAGuy

Frankly, this thing has no business as a standard of any kind. The grammar may be allegedly simple, but the semantics of the stuff inside there are incomprehensible to anything other than Microsoft Office.

The option at that time for Microsoft is to fix the errors, axe the bloat, and come back with something that IS suitable… like extensions to ODF. That way, high standards are maintained and sanity prevails. What are the odds?

Hi Anonymous #2, Standards reviews have their own form rules and procedures. These are very detailed and precise and are intended to ensure that the final standard product is also precise.

A standard is not formally defective just because it is missing a feature that some users may want. The standard may fail in the market because of that, but it is not formally defective in ISO. On the other hand, a specification of great length and detail can be formally defective because of procedural or drafting errors.

Rob – Great article. The only problem I have is that I don’t know how to follow the actual proceeedings except through commentary elsewhere. Has JTC1 really “accepted” the Microsoft argument, or is that just Brian Jones playing to the court of public opinion. It is not absolutely necessary to define deviancy down. Instead, would it not be possible to simply take a sufficient number of errors and declare the standard void? Perhaps not, due to the idea that you have to have a way to turn a No vote into a Yes vote, but I don’t know. Could you elaborate on the process and why the fast-track ballot could not simply find “enough” problems without having to enumerate all of them?

Gary – Hot damn! Have at ‘em. Much as I like and admire Rob’s way with language, a bit of “through the heart” activism sounds good too.

- Ben

Picture a carnival cruise ship in a standard olympic sized swimming pool. It WILL scrape the bottom. It won’t be able to move.

I’m not quite sure what “fast track” is specifically for, but I’ve dealt with lots of standards (SAE, IEEE, RFCs) that deal with some very specific function, so in the dozens of pages are easy to analyze. I was also involved (including writing a reference implementation) of RFC 2440 – open PGP, and there was a LOT of discussion and refinement even though the technology and “simple things” like formats was thought to be well understood. Fast track is probably for something where there is already a body of interoperable implementations that all play by the same rules which are well known and there isn’t much debate about and just needs the rubber stamp.

Fast tracking 6000+ pages that might or might not document one completely new implementation properly is absurd. It would be somewhat less absurd if everything external was also standard (e.g. SVG instead of VML).

And what is the purpose of making OOXML a standard anyway? Here I don’t care if it competes with ODF or not, merely if it will truly be an interchange standard so that anyone can have a shot at creating an interoperable program. And that is where the “contradictions” become important. You can’t code for something to do one kind of spacing and a completely different kind of spacing at the same time at the same place in the same document.

And I’ve already voiced my doubts – why bother with OOXML at all when nothing that even gets the 6000+ pages perfectly implemented (assume the programmer is psychic and will know how to resolve the “contradictions”) will be able to properly render what is emitted from Office 2007?

Microsoft wants the ISO stamp of approval so they can call their software “open” and “standard” – which will be defined downward if this goes through.

Microsoft could simply support ODF.

They could have proposed extensions during the long process of standardizing ODF to make it easy – and they were on the committee.

They can propose an extension NOW to ODF to add whatever they think is missing between Office 2007 and the current ODF standard and it would not run 6000 pages, and might be something that could be fast-tracked legitimately.

And don’t assume Microsoft properly documented Office 2007 in OOXML – if they made a mistake, and it becomes “standard”, what will they do? Will they correct Office 2007 so it will obey the specification it submitted (or leave it “wrong” like Internet Explorer, and assume everyone will make their implementations compatible with the errors)?

So when Microsoft Office 20xx supports neither (the current fast tracked or even a fixed) OOXML nor ODF properly what then?

However so far there is only a few issues found that really warrent any action. And none even that can’t be corrected in a 1.1 version of the spec simular to ODF that is still correcting flaws it it’s ISO specification for future versions.

I was happy to read
http://tirania.org/blog/archive/2007/Jan-30.html
though as it shows even OSS experts are divided a bit in support for OOXML as ISO format.

Still, it’s a bit misleading to think of OpenXML’s 6,000 pages as one format. It really specifies three formats for totally different applications (word processing, spreadsheets, and presentation.) Given the complexity of these applications, I do not see how the format could be reduced by more than perhaps 1,000 pages.

If you have found problems in the OpenXML spec, please post them online–maybe as a marked-up Word or PDF file.

Sadly I’m not so kind to ANSI/INCiTS for letting Microsoft skate through ISO/IEC without proper review and due diligence. IMHO, it’s a national outrage and a cause for shame. But then, i’m coarse and cynical; suspecting undue influence and the ever present beauracratic willingness to accomodate.

That deviance thing? Can i really get rid of my tin foil hat if i learn to talk like that?

Still, i wonder if others see what i do. You have the sharpest dagger, and the most dangerous, if not noble of intentions. The difference is that you cloak your thrust under the guise of such courtly manners and erudite discussion, perhaps the bastardos will not see it coming. Maybe. For sure they don’t know how to handle your assault.

For me though, it’s through the heart. There is no excuse for the position ANSI/INCiTS has taken on the Ecma 376 fast track ballot.

If it looks like a duck, quacks like a duck, walks like a duck, and squirts crap like a duck, then ANSI/INCiTS is indeed the corrupt and Microsoft accommodating organization we suspect it to be.

That they would have adopted a definition for “contradiction” suggested by Microsoft and serving Microsoft’s needs is shameless. I’m embarrassed to be an America.

~ge~
Yankees in the Court of King Arthur, with a Microsoft Agenda

Since you apparently found an additional 20 problems in the Ecma 376 draft, it would be great if you could add them at the EOOXML Objections Clearinghouse wiki for future reference. If you don’t have time to write them up carefully, it would still be great if you could jot a few quick notes at the Talk (discussion) page.