Why I Oppose Microsoft’s OOXML
It was recently announced that the ISO has approved a draft standard based on Microsoft's OOXML format, which was developed for Microsoft Office 2007 as a smaller alternative to the original binary-format files used in previous versions of the suite.
I am personally opposed to Microsoft's format becoming a standard. That's a simple fact. The reasoning is more complex, and has (I freely admit) been influenced over the last several months by articles I have read on the subject.
A good deal of those articles were (and are) hosted at NoOOXML, which also hosts a petition against the standard. The issues with the standard are numerous; I won't try to cover here what the great material at NoOOXML has already put so eloquently. All I'm trying to do is raise awareness of the issue.
The first bone comes in the shape of a preexisting standard, known as ODF, or Open Document Format. ODF is another XML-based format that was approved by the ISO and used in products like OpenOffice.org. Open-source stuff.
Microsoft noticed that businesses and governments around the world were starting to switch from its proprietary formats to the open standard. The file sizes were smaller, etc. etc. Microsoft decided to create its own XML-based standard for Office 2007, rather than using the existing one.
At this point, it is important to note the differences between the two formats. Courtesy of NoOOXML, here's a short list of problems with Microsoft's format, links preserved for easy info:
- Wrong goal
- Binary Space
- 1900 bug
- Patents
- Accessibility
- Global relevance
- User needs
- Open Standards
- Conversion Issues
- Rice Pudding
- Only Latin URLs
A couple of those aren't really obvious, so let me clarify. "Rice Pudding" refers to the length of the standard and the indications that ECMA didn't do a proper review of the standard. "Conversion Issues" refers to the lack of backward-compatibility with the existing ODF standard. "1900 bug" is a reference to the bug in Excel (and therefore OOXML) that causes January 1, 1900, to be reported as Sunday, while it was in fact a Monday.
Anyway, moving on from the list (I think the rest are pretty self-explanatory), there is also some contention about why OOXML was fast-tracked through the approval process. Rumors that Microsoft "stuffed" voting in some countries to guarantee approval by that committee, and other unpleasantness. Regardless of that, there are some other points to be brought up.
The OOXML standard is supposedly written to be fully implementable only on the Windows platform, meaning that Linux and Macintosh users would be unable to use the formats. Users who adopt the OOXML file format would be stuck on Windows from day one.
Another issue brought up on the NoOOXML site is the binary data integration, which Microsoft supposedly uses for "backwards compatibility" purposes. The details of those binary chunks are not published for third parties, making Microsoft the only developer with full knowledge of how to utilize the feature.
I could sit here typing out arguments for hours, but I'll let you head to NoOOXML for more information if you want it. From here, my post will turn into (even more) pure opinion.
Above arguments aside, I don't see that there has been any input from other companies or organizations. Microsoft has been the sole developer of the OOXML standard. In contrast, the ODF specifications were initially developed by Sun Microsystems, then development was taken over by the Open Office XML technical committee of the Organization for the Advancement of Structured Information Standards (OASIS) consortium. To my knowledge, there are no patents on ODF; Microsoft can charge for licenses to implement OOXML.
As I've said already, the list goes on and on. I'm keeping the badge I have in my sidebar for the time being. Perhaps I'll even move that module up higher on the page (with the other things promoted in it) to raise awareness.
My sentiments are perhaps best summed up by this quote from my friend i80and, who posted it to Twitter:
Aw, man! OOXML has been accepted as a standard!
Indeed, that was the final straw that caused this blog post's list entry to appear in my topic queue. (He obviously didn't tag the acronym; I did that for this post.)
I wonder if writing this a few months earlier would have had any noticeable effect. Fortunately there's still two months for an appeal.
So what are your opinions on the OOXML controversy? I'd love to hear them in the comments!
Google Infringing on Sun Patents with Android’s Dalvik?
OK, so Google doesn't call it a Java VM; to them, it's "Dalvik", not Java. It does accept Java bytecode, though, and turns it into Dalvik bytecode. Sun hasn't filed any lawsuits yet, and as far as I know doesn't consider Google's custom virtual machine to be a problem or infringement on their intellectual property. But the developer community and Sun are going over the nonstandard VM that Google made.
When Android came out, I was determined to ignore it, as mobile phone technology doesn't really affect me these days (not until I actually get a cell phone). Since then, Google's announced a full SDK, $10 million developer competition, and shown prototypes that look pretty promising, actually. OK, so all I saw of the prototypes was a few pictures on another blog, but it looked cool. And I can't ignore the Linux part. Not anymore, at least.
Anyway, some people call Google's move "clever", among other things, mostly positive from what I've read. That word above is the opinion of Stefano Mazzocchi, a developer and board member at Apache Labs. Others, such as Hari Gottipatti (a mobile developer), say Google converts the bytecodes so it can say Dalvik is not a Java virtual machine. The general consensus, at least according to my take on the article I read, seems to be that Google is somehow avoiding Sun licensing fees, though (says Gottipatti) Sun might have just waived the licensing if Google had simply approached them and asked.
Me, I'm not too terribly interested in the whole Android thing, but this particular part of it, the maybe-Java-maybe-not part, seems to be worth following. Next thing is to find out whether or not Sun will pursue legal action against Google for what sounds like, on the surface, copying Java Micro Edition. Such a lawsuit probably wouldn't be a good vibe for a company so active in the open source movement, but you never know. I'll definitely keep an eye out for further developments on this one; it's piqued my interest.
PS: To anyone who's ever used the phrase "peaked my interest" or "peeked my interest" (ouch!), please go to your blog control panel right now and fix it. Now! Reading those two words in place of the real one hurts my brain; they give a different (and often nonsensical) meaning to the sentence. How do you peek an interest? I didn't think so...



