When it comes to standards, agreement to follow the standard is as important, sometimes more important than what the standard is. Width of twin beds, or diameters of hose pipes etc, half in inch? or 12 mm? INBD. In more complex products, be it engine lubricant temperature vs viscosity profile or HDTV protocol, we need some impartial body which has no dog in the race to set the standard. SAE for lubricants, ACM for ASCII codes, or IEEE for communication protocols. Further the standards should be free, as in beer, unburdened by IP/patent liability.
What Microsoft did in OO-XML was to encapsulate its old binary data files inside XML tags. Some of the old Word file formats do not have any explicitly defined structure or behavior. If you feed that binary data to old Word executable, what it does is what the standard is. No one else can implement that data without having the actual old Word binary executables. Even if Microsoft agrees to release all the old binary executables for all to use, it still would not run in Linux. Microsoft has lost its original source code, and it could not be recompiled in linux, it looks like. Even Microsoft is not able to get a fully compliant MsOffice for Linux.
It is completely to wrong to claim OO-XML is a "standard" in any definition of "standard" we engineers have come to understand. But Microsoft muddied the well, employed very articulate lawyers to confuse the issue. It was all pointy haired bosses and lawyer talk.
Those days it was difficult to believe that Microsoft could be tamed. But eventually it was