A while back the OpenDocument Foundation folded up, withdrawing its support for the ODF in favor of CDF. The reason for the switch is buried in the details of ODF community’s denial to be fully interoperable with Microsoft Office, which might have helped in migrating to ODF without affecting the processes.
So, there was something bigger here playing it up. Matt Assay notes that Microsoft Sharepoint might kill ODF more than anything else. It is the process stupid! People want to move to ODF, but without having to re-engineer their business processes. And if their processes are being controlled by a closed proprietary tool, it is going to be increasingly difficult to build support for ODF in it. Especially if it already has support for the other format from the same vendor. In a nutshell, OOXML might win over ODF not because of its merit but because of Sharepoint’s support for it.
It is true that the office documents are only a part of the ecosystem, and choice of their formats is going to be dictated by decisions for rest of the ecosystem. Unfortunately many of these businesses might have already invested in proprietary tools that run this ecosystem. The ODF community has to realize that tools that will help a business migrate within this ecosystem are important to promote the format. And this migration might involve a transition period of co-existence of both the formats and their tools.
But I wonder if the solution is to modify ODF to only to make it fully compatible with Microsoft formats. Maybe and just maybe it might be worth it to use ODF derivative during the migration, with plugins for it. But the key really will be to minimize the transition time. I can imagine that it will be difficult for people if one fine day a software gets replaced by a brand new one on all the machines. But it would not be so difficult it they knew about this a month before and were being trained for it.