Here, you have the links with the modern hebrew transliteration of those roman coins. That´s one heterodoxe way to read and know their inscriptions. I think we are misreading and confusing ancients.

I have been reading on the Denmark situation on Updegrove’s blog. It is pretty obvious the policy makers don’t get that OOXML is not suitable to deliver the benefits of a standard. It looks like if it is labeled as a standard and submitted to standard bodies like ECMA and ISO, then it is assumed that it is a standard and will operate as one. It’s like a reverse “duck test”. When it is called a duck, it must a duck that walk like a duck and quack like a duck.

There is a lot of competent information to the contrary but it is not easily accessible outside the document expert community. This information is sprinkled throughout the blogosphere and it requires time and dedication to retrieve it.

I do not want to single Stephane’s work here because the blogosphere is by nature scattered. However I can’t help but notice this excellent information is among the hardest to get at. It is available when you know where to look, but one has to jump through a lot of hoops to find it.

I think there is a need for a single place where all the reasons why this duck doesn’t qualk and walk the way it should are stated in plain English in terms easily understandable to a non-expert policy maker with pointers to substantiating evidence. Then the policy makers can ask their experts to see for themselves if the claims are true or hype.

A wiki would seem the best tool. As new arguments are developed in the blogosphere, anyone would be able to add them to the central refenrece place. Perhaps Grokdoc would be appropriate.

Not all of us like understatement, minimalism, Apple design, or whatever it is called these days. Cathedrals, Victorian furnishings, and rain forests are striking precisely because they are so ornate and overladen.

(Granted, there is merit in keeping standards and instructions concise.)

OOXML extensions by themselves don’t guarantee any more than what ODF does. A vendor could just define a schema extension that allows name/value string pairs. You can’t legislate good taste.

In any case, I’d also note that ODF allows a vendor to add extensions as well, with a vendor-defined schema, if they wish. No one forces them to use settings.xml.

Honestly, if you examine OOXML carefully, it isn’t exactly a model of a tight schema that allows meaningful validation. Most of the types are merely strings without additional constraints. So it is difficult to argue that the schema is a significant source of interoperability. That, plus the fact that Ecma failed to submit the OOXML schema to ISO for approval.

The ODF config items have no origin, cannot be validated and are designed to be inoperable by default.
Remember that ODF has interoperability as one of it’s main goals. For OOXML compatibility is a more important goal than interoperability.

There are my comments at INCITS JTC-1,

http://www.incits.org/DIS29500/DIS29500.htm

(at the moment, the site is down apparently)

Some of these are simply devastating when you think in either :
- backwards compatibility
- full documentation
- interoperability
- completeness
- ability to compete

And of course this article itself,

http://www.codeproject.com/useritems/office2007bin.asp

You know, what does not trump reality is that I, as an independent vendor, been doing this work (of implementing both old Excel BIFF files, and new Excel 2007 files) for some time now. When you have nothing more than a new binary file format surrounded by angle brackets, it takes as much time to reverse engineer than what was never documented with BIFF. It’s fire and motion from Microsoft, the same old game.

In fact, I posit the situation for non-Microsoft vendors right now is the situation with no ECMA 376 at all. For any real and substantial implementation, we are on our own, back to reverse engineering, back to spending years to decipher this new beast.

-Stephane Rodriguez
http://xlsgen.arstdesign.com

You sprinkle lots of wisdom in bits and pieces as comments into other people’s blogs. I enjoy your posts greatly. Did you consider posting some of the key facts in an article that can be linked to and used as a reference?

There are lots of organizations that are now pondering whether they go ODF or OOXML. The claim of backward compatibility is a big part of the equation. They need to know that existing binary Office document won’t end up in usable XML once converted to OOXML because of the bin parts. Any chance the same document gets into proper XML if the target format is ODF?

About the VMWare off-topic, I fully expect these tactics to backfire severely. People has wisened up to Microsoft and they see them coming miles away. “We must buy Microsoft or they will break our infrastructure” makes poor purchase business cases and breaks the confidence customers have into their supplier. There are also limit to how many third party vendors they can kill without seeing the others taking defensive measures and they have crossed that line a long time ago.

That takes me back. Anyone else remember DR DOS, and how Microsoft added code to the beta of Windows 3.1 to prevent Windows from being installed if someone had DR DOS rather than MS DOS? Similar issue with PC-DOS and Windows 95. When will they learn?

My prediction is this all ends with a “Linux for Windows” which will be the only Linux that works in Microsoft’s virtualization environment. Only blessed commercial distros will be supported, and Microsoft will collect a royalty on every copy sold.

Although it has nothing to do with OOXML, here is a sobering read about Microsoft predatory practices, and stifling of innovation, in the virtualization space.
http://www.vmware.com/solutions/whitepapers/msoft_licensing_wp.html

I think there are parallels with what Microsoft is doing with OOXML (in particular OOXML is only a piece of the integrated Office suite that is designed to make it impossible to compete for anybody)

– offtopic alert –

-Stephane Rodriguez

In addition, contrary to older MS Office files, there is a new binding mechanism between VBA macros, OLE objects and the XML parts (attribute “CodeName”).

None of the above is documented.

-Stephane Rodriguez
http://xlsgen.arstdesign.com

My article over at codeproject is only partially about BIFF12. It so happens that many of the binary parts are written as .bin ZIP entries, be it VBA macros, OLE objects, or BIFF12 parts.

So there is no mistake, what we are talking about here are .bin parts which are integral parts of OOXML. Without .bin parts, it is impossible to successfully migrate older MS Office files over to MS Office 2007.

-Stephane Rodriguez
http://xlsgen.arstdesign.com

Irreducible complexity — I admit that term isn’t part of my working vocabulary. But I like it. Reminds me of Saint-Exupéry’s line, “Perfection is achieved, not when there is nothing more to add, but when there is nothing left to take away.” My personal design aesthetic is very much along those lines. It doesn’t mean that you do less, or have fewer features. It just means that you achieve more with less.

Or said different, William Strunk on writing: “Vigorous writing is concise. A sentence should contain no unnecessary words, a paragraph no unnecessary sentences, for the same reason that a drawing should have no unnecessary lines and a machine no unnecessary parts. This requires not that the writer make all his sentences short, or that he avoid all detail and treat his subjects only in outline, but that every word tell.”

Of course, my main blog post was far from concise, though in defense, I try to entertain as much as inform.

Hi Stephane, you may have a point there with the Office 2007 BIFF formats, but those are not being proposed as standards. However, if there is a parallel issue with the OOXML formats that has not already been mentioned, then you should elaborate on that.

This alone contradicts their “fully-documented” statements, don’t you think?

You can go on record with the quote above.

-Stephane Rodriguez
http://xlsgen.arstdesign.com

“A system, device or construct is irreducibly complex if it has a number of different components that work together to accomplish the task of the system, and if you were to remove one of the components, the system would no longer function”, (Author Michael Behe, “Darwin’s Black Box”).

I’ve always thought the XML 1.0 specification to be a work of “irreducible complexity” in that they were driven to define things to the point where anything less would have broken the concept, yet there was infinite headroom for innovation.

A driving objective for ODF XML has been to become an application and platform independent universal file format that can be implemented across many different profiles while maintaining a higher quality of interoperability. It was designed as a portable document/data container to transition across sprawling information processing chains where applications of many purposes span desktop, devices, servers and Internet systems.

The presence of application specific settings as named values in the core of the document content or presentation layers is a killer when it comes to transitioning a document/data container across different applications. Unless of course the only applications in the processing chain are certified by our friends in Redmond.

I guess there is a method to their madness.

~ge~