I agree with almost everyone that it's highly unlikely that Oracle will join up, although it is possible. Oracle has very little interest in anything that doesn't contribute to the price of fuel for one of Larry Ellison's jets. That means that Oracle's main focus will be its server software stack.
Ellison does have some interest in OpenOffice. Back in June 2009, not long after Oracle purchased Sun, Ellison proposed that OpenOffice developers quit using C++ for OpenOffice and switch to using JavaFX. That idea went over like a lead balloon.
Even before Ellison proposed that developers dump their ten-plus years work for an unpopular language, OpenOffice programmers weren't happy. Getting new features or just bug-fixes through the OpenOffice organization was a long and painful process, whether it was controlled by Sun or by Oracle.
In general, Sun's top developers haven't been happy with Oracle anyway. James Gosling, Java's creator, left saying that Oracle had been low-balling key employees and cutting projects off at the knees. A who's who of top programmers could be made from former Sun employees alone. For example, besides Gosling, Jeff Bonwick, creator of ZFS, the popular high-end file system and logical volume manager, has just left Oracle.
The long and short of it is that Oracle is not a place where the best developers want to work. Many of the software projects Oracle inherited from Sun still have their devoted programmer fans. OpenSolaris, which has been left to die by Oracle, now lives on in Illumos. I see the Document Foundation as a preemptive strike to keep OpenOffice alive in its current form as a standalone desktop application.
What's happening now is that LibreOffice will come out in a final version by year's end (it's already available in beta). Its developers tell me that first version will consist mostly of long-delayed bug fixes and a few new features that are already available in Novell's version of OpenOffice, such as the ability to read and write Microsoft OpenXML formatted documents.
After that, what happens with LibreOffice will all depend on what the developers and end-users want -- not what Oracle wants. If Oracle doesn't like it, then we'll see a fork.
What I mean by a fork, by the by, is an actual split in the code. For example, Ubuntu can be seen as a fork of Debian. No one doubts that Ubuntu is based on Debian Linux, but it's also clearly a Linux distribution in its own right. Simply changing out some trademarks and product names, which, for example, is what Oracle did with Red Hat Enterprise Linux when it created Oracle Linux, isn't the same thing. At this early point, that's all the Document Foundation has done with OpenOffice.
My expectation is that Oracle will quietly let OpenOffice gather dust, and LibreOffice will become the new open-source office suite of choice. What do you think?