OOXML and ODF
You may wonder why OOXML and ODF are such big issues. After all, they are both zipped XML-based storage formats. Why couldn't someone use XSLT to transform documents from one format to the other? Isn't that one of the chief advantages of XML?
There are some technical obstacles, albeit they may not be insurmountable. The biggest technical problems stem from the over-specificity of the OOXML format. For example, OOXML specifies the cute little border graphics that are sometimes used around documents, giving each one a number. What if you want to use a different graphic in your implementation? The ODF standard says to enclose a copy of the graphic inside the zip archive and link to it from the XML file, which makes the receiver's software able to render said graphic even if it is not the same software as the sender uses. On the other hand, OOXML, with its pre-specified graphics could mean that every application needs to have its own copies of those graphics. Those graphics are copyrighted, so a competitor implementing OOXML would have to get permission to use them, but they are built into the standard as a requirement.
It is difficult enough sometimes to find ways to utilize the data files of a competing product, with its "magic numbers" and built-in assumptions, but as I mentioned here, OOXML has also been developed as a "memory dump" of Microsoft's suite of office applications, which means that the import and export process must reproduce the internal conditions of Microsoft's software, including binary data inserted within the XML files. You have to wonder there, what in the world were they thinking?
Because OOXML is designed to mimic one existing software suite, it retains all of the errors that were made during the creation of said software, such as the Excel bug in which dates are off by one before a certain date. This is mentioned in a fact sheet put out by the ODF Alliance, as is the binary blobs issue. Also, there are already standards for XML representations of things like graphs (SVG) and mathematical equations (MathML), so why in the world doesn't OOXML use those standards like ODF does?
Another issue, and a very important one, is the forest of patents and patent applications that Microsoft has placed around OOXML.
But one of the clearest expressions of what this dispute is all about is found on GullFOSS. If the whole purpose of ODF is to make it easily possible for anyone, using any office suite or other application that they may have, to use the data files it produces, why in the world would someone use any other format? This is why Google Docs, OpenOffice.org , KOffice, and AbiWord all support ODF. ThinkFree Office is said to be releasing full ODF support in their next version (expected release a few days away), while WordPerfect is releasing partial support in their next version in 2007. The OOXML format? It is doubtful that they had the time or the freedom to consider how well it would work with other applications.
The nail in the coffin should be this comment , "end-users will have greater control over their data when saved in ODF." It is for that reason that I believe that every federal, state, and local agency should be required to use ODF for their editable data. Anything else is just plain wrong. Public data should be available and accessible for the entire public body, even if they do not have the funds to buy a particular office suite, or they use a different operating system than a particular office suite supports.