To read all comments associated with this story, please click here.
OOXML is suppose to be is little more than a thin wrapper around a binary dump of the in memory format used by Office
This would be fine by itself - as long as each and every possible component included in the format would be fully documented and unencumbered. As far as pure basic structure goes, OOXML is OK.
Although I'm no expert, I think Rob (from IBM) has raised several valid criticisms of OOXML (such as the use of bit fields in XML and incorporating date hacks into the standard):
He conveniently forgets to mention that the "date hack" originated with Lotus 1-2-3, and was adopted by Excel and subsequently every other spreadsheet application that wanted to be compatible with the many Lotus 1-2-3 documents that had been/were being produced. It would be more work (and error-prone) to create a new date representation and transform every Lotus and Lotus date-compatible spreadsheet format not only for the new format but also for the internal representation of many existing applications.
The bitmasks define unicode code pages and are documented in the spec. Neither tie the format to Windows nor present barriers to implementation.
//He conveniently forgets to mention that the "date hack" originated with Lotus 1-2-3, and was adopted by Excel and subsequently every other spreadsheet application that wanted to be compatible with the many Lotus 1-2-3 documents that had been/were being produced. //
You are trying to make it sound like Microsoft support compatibility. Nothing could be further from the truth here.
If Microsoft were truly out to achieve compatibility and interoperability, then why don't they just adopt the open standard formats that are designed to work on every platform?
There are a whole raft of these. SVG, Ogg, ODF, SMIL, Python Perl Java or Ruby, full PNG, full CSS ... and so on and so on. All open standards, all free to be implemented by any party, all unencumbered so no-one gets sued, and all of them promoting true croos-platform interoperability.
Why is it that in every single case, Microsoft refuses to support the perfectly viable suitable and available open standard, but instead tries to insist that its own proprietary, close, patented, single-platform only equivalent format is the only one Microsoft will support?
When Microsoft and Novell announce a deal to improve interoperability, why is it that the avenue to that turns out to be Microsoft's always-proprietary-lower-level-standard-OpenXML format to be implemented in OpenOffice and not the open-all-the-way-down ODF format implemeted instead in MS Office?
When will we see Microsoft announce MS Office for Linux?
Can you say "lock-in"?







Member since:
2006-08-21
Although I'm no expert, I think Rob (from IBM) has raised several valid criticisms of OOXML (such as the use of bit fields in XML and incorporating date hacks into the standard):
http://www.robweir.com/blog/labels/OOXML.html
Reading his articles left me feeling like the "new and improved interoperability format" that OOXML is suppose to be is little more than a thin wrapper around a binary dump of the in memory format used by Office.
If this is true, it does not strike me as a good thing (talk about defeating the purpose of XML).