I’ll be offline for the next week or so. Sorry if I don’t answer your e-mails or comments during that time. I’ll try to get caught up early August.
In the mean time though, check out this article up on eweek: Setting a New Standard
As a user of OpenOffice.org on Linux who works in a mostly Microsoft-formatted world, I’m somewhat of a stakeholder in the ODF-vs.-OOXML horse race, and I’d like to see OpenOffice.org take advantage of this marginal boost in openness. In particular, I’d like to see vendors and projects that back ODF and OpenOffice.org attack the OOXML spec’s 6,000 pages that Microsoft has offered to standardization bodies and do so with less focus on teasing out ISO inadequacies and more on identifying methods for improving support for Microsoft’s legacy Office formats.
I don’t necessarily agree with everything Jason has to say, but it’s still a good read. There are certain things I like about ODF, but that doesn’t mean it’s the format I’d choose if I were building an application from scratch. It would really depend on the application. For example, if I were building a spreadsheet app that was going to handle large amounts of data, the features in spreadsheetML around splitting each worksheet out into its own part; shared string tables; etc. would be super valuable (see this post for more info). If I just wanted to do basic table manipulation though, then the ODF model where a spreadsheet is similar to an HTML table might be preferable.
The same is the case for documents and presentations. It’s really just dependent on what kind of model I’m using for my app and how I want it to behave. Either format would work in many cases, but there’s probably always one that’s more preferable (and it’s not the same one for all applications).