I appreciate the fact that Jonathan Schwartz of SUN has jumped into the doc format discussion with a well thought out discussion. I think there are some important issues raised, control of your data, long-term archival of data, and most importantly, the translation of documents.  Yet, his posting follows some very careful positioning in order to lead to the logical conclusion (for him) that you should use his products and services. Logical - but it is important to keep an eye on the issues rather than the commercial interests.

Yes - my employer has commercial objectives to see our product used. Thus goes the nature of competition.

 1) Control of your data is critical. We certainly recognize that - thus the move by most formats to being XML-based. The base level standard is XML, and that gives enormous opportunity for access to the data no matter what application created it, or when. 100 years from now, XML will be the magic bullet, not ODF or Open XML. 

2) Long-term archival is a huge issue and challenge. There is no question that the accessability to the Declaration of Independence is not based on a requirement for a machine to be able to read the document. Knowledge of the language sure helps though. The trade-offs in the past for data storage vs. computing capability were such that it was not feasable to have the benefits of XML-based formats. How much of the flat data stored on mainframes is readable without the mainframe applications? How about the data stored in directory services or other data manipulation systems.

I have never found the archival arguments to be complete. If you were to go completely down this path then SAP would have to open all of its data, Oracle, IBM (DB2), etc. etc. The trade off that everyone is ok with is based on the benefits that the system may offer (value) in exchange for the difficulties with long-term archival (or any one of a hundred other trade-offs).

Microsoft responded to national libraries and government agencies years ago with access to the binary formats, and source code for the Save As functions etc. Also, we provided documentation and licensing that recognized the import of the long-term archival issues. Is that a perfect solution - nope, but it was a step in the right direction. ODF is no better nor worse than Open XML at enabling technical access in 100 years simply because both are XML files and can be opened up easily. In fact, the Open XML translator project shows that it is a relatively easy task to build an independent piece of code that lets you do that.

 3) The point that Mr. Schwartz makes about the bridge that is being built between Google formats and OpenOffice is exactly what we have been saying for over a year. Translation is the key. There are dozens of document format standards, some from consortia, some from national bodies, and some from international bodies. The whole "only one" argument is really a commercially-driven sentiment that does not reflect the market reality, nor the desire of customers to have more choice - not less. I think it is great they are building translation capability, I hope they continue to do so as it is the exact right thing to enable for customers.

 Interesting letter from Mr. Schwartz - glad the conversation continues.