Ecma Standard 376 – Office Open XML formats

It's finally official. Today the Ecma General Assembly voted almost unanimously to approve the Office Open XML formats as an official Ecma standard. They also voted to submit the standard to ISO for fast track certification. The official press release from Ecma International can be found here:

Here's a quote from Jan van den Beld, the Secretary General of Ecma International:

"The broad spectrum of sponsors from the industry and public institutions ensure the creation of an open standard that can create a wide range of possibilities for document processing, archival and interoperability" said Jan van den Beld, Secretary General of Ecma International. "The Open XML standard recognizes the benefit of backward compatibility preservation of the billions of documents that have already been created while enabling new future applications of document technology."

As I blogged about back in early October when we finalized the draft, it was a ton of hard work by around 20+ individuals from about 12 corporations, but it was well worth it. The dramatic improvements that the specification has undergone over the past year show it all. I was just looking back at one of my earlier blog posts from last year when we first submitted the Open XML formats to Ecma for standardization. It really is amazing that the group was able to accomplish so much work this past year. I can't wait to see what we're able to accomplish as we start working on the next version.

There are a few things that will be pretty fun to watch over the next few months. The one I'm most excited is the huge growth we're seeing in the Open XML developer community. We already have larger corporations/applications like Apple, Corel, Novel, Intel and Microsoft either participate directly in the standardization, or publicly comment on upcoming support for the Open XML formats. This is cool, but not the piece I'm as excited about. I'm looking forward to all the smaller companies (like the folks participating up on that take advantage of this interoperable format as a building block for rich solutions that create and consume documents, spreadsheets, presentations, and other document types we haven't thought of yet. There are simple tools out there like the docx converter ( There is the open source project up on sourceforge for converting from Open XML to ODF and back ( There is even the MindManager tool that converts mind maps into wordprocessing documents. The possibility are infinite… who knows how long it will be before we even see things like rich spreadsheetML files with branding, charting, pivot tables, etc. being output directly from hardware devices (like medical imaging equipment) rather than a basic CSV file.

Next step is going to be ISO certification, and here's hoping that IBM doesn't try to push too hard against this too. They've already said they were the only ones who voted against the Ecma certification which is a shame. Having a complete spec that outlines every piece of the Office file formats and is publicly maintained is a good thing. We shouldn't have to waste time arguing about that (unless we're in training to be a politician or something). Open XML may not be the best format for every use, but what is? HTML is great for some things, but really limited for others. XML is great for some things, but really limited to others. The same goes for these newer formats like ODF an Open XML. Microsoft had no problem with ODF becoming an ISO standard for instance, and didn't put up any roadblocks. It was a good thing. Hopefully IBM will be able to say the same about Open XML, but we'll have to wait and see. I haven't seen any public comments on that yet.

So, to all of my colleagues from TC45, congratulations! I think these stats say it all in terms of your hard work:

  • 72 presentations were given to the technical committee explaining the existing behaviors of features so that discussion on how to best structure and document it could then take place.
  • 66 hours of live meeting discussions (starting at 6am every Thursday for those of us on the west coast of the US)
  • 88 schema files
  • 128 hours of face to face meetings held in Brussels (ECMA); Cupertino, CA (Apple); London (British Library); Sapporro, Japan (Toshiba); Redmond, WA (Microsoft); Trondheim, Norway (StatOil)
  • 6,000 pages of documentation between the 5 parts of the standard
  • 9,422 different items to document (3,114 attributes, 2,500 elements, 3,243 enumerations, 567 simple types)