Open XML links for 12-13-2007

The last week has seen some interesting discussions and useful how-to posts on Open XML blogs …

Three new articles on Sheela over at Sonata has recently posted several new articles, including an overview of how digital signatures work in WordprocessingML, information about how to implement headers and footers in SpreadsheetML, and the first in a two-part series on Open XML’s implementation of mail merge in WordprocessingML.

Eric White can’t stop posting Linq to XML samples. He’s at it again, this time with a class for WordprocessingML documents. Man, I can’t help wondering what his pace will be like after he actually starts the Open XML job next month. (Note to Eric: before you get too carried away, I’ve heard from one of your current teammates that they still need you to finish up a couple things. :-))

Standards groups meet in Kyoto. TC45 met in Kyoto last week, and reviewed a large number of proposed dispositions for Open XML comments, which led to over 1000 comments getting new proposed dispositions for review by the ISO national bodies. This week another standards group (SC34) met in Kyoto, and Rick Jelliffe’s blog offers a few observations from a participant.

I was disappointed to learn in the comment thread on Rick’s post that he won’t be attending the BRM for Open XML. As he explains,

“Standards Australia and I discussed it informally, and I suggested that I might be a bit too controversial a choice to go. IBM wrote a letter of complaint about me to Standards Australia, after I spoke on the history of SC34 and various issues about standards (at the invitation of SA) earlier in the year, and I don’t want to expose the SA officers to more of that kind of trivial nastiness.”

Service Pack 1 is now available for Office 2007. Brian Jones has a post on SP1 that describes one change that allows for improved integration with document-format translators.

Brian’s previous post on the latest batch of comment responses from Ecma (which I linked to earlier this week) has prompted some interesting comments that are worth checking out.

There is also an interesting thread on Rob Weir’s “Bait and Switch” post about the proposals for maintenance of Open XML. I found this exchange in the comments section telling:


I need the file format of NSF-files. I want to decrypt properitary encap2.ond attachments (they are tiny NSF-databases and the equivalent of TNEF for MS-Office which has been reverse engineered very well) on an embedded linux system and I cannot place notes.dll or the whole Notes client onto that appliance because of licensing issues.

Rob’s response:

encap2.ond attachments? They haven’t been used since Notes 4.5, like 10 years ago. How many of those do you still have?

In any case I’m not aware of any NSF file format specification. Your best bet may be to look for a way of doing a one-time programmatic migration of the data on a machine that does have Notes installed, into a format that you can better consume.

Documentation of the details of legacy documents apparently isn’t as important as it once was. Or is the NSF issue different because the solution requires installing a closed-source proprietary IBM product?

Comments (10)

  1. Rob Weir says:

    Well, duh, one obvious difference is no one is proposing NSF format as an ISO standard.

  2. Alex says:

    Rob, whether you duh or duhn’t, a lot of people suggest opening MSO legacy formats (not that they are not well documented already). Double standards FTW? 😉

  3. Rob Weir says:

    Well, Microsoft once did have public documentation on their formats, then Bill Gates wrote in an internal memo that was "crazy" to share the documentation with their competitors, and then the documentation disappeared for years, only to reemerge when Microsoft wanted to appear open for ISO.

    So I think there is a difference between a file type that has no documentation, internal or public, and one that Microsoft has historically used as an anti-competitive control point.

    Note also that Microsoft’s claim for OOXML is that it is valuable because it is 100% compatible with the legacy formats. But this benefit is clearly not achievable by anyone unless they have the legacy file format documentation, or continue to run MS Office.  Access to the legacy formats is 1/2 of the story here.

    So because of the fact that IBM is not trying to convince ISO to standardize a format that is 100% compatible with Notes NSF formats, your analogy doesn’t go far.

  4. Alex says:

    Rob, for all I see ISBN 1572314982 is still available for those who want to work with BIFF8 – the most recent legacy binary Excel file format. You cannot really pull a book once it’s out in the wild, you know 😉 And although this book is incomplete and has errors, it was and still is available for 10 years now. What do you have on NSF formats anyway?

  5. Rob Weir says:

    Indeed, I have a copy of that book….someplace.  And for the Word or PowerPoint formats I suppose you can get some of that info off of old MSDN CD’s, circa 1998.  But that is rather old, don’t you think?  Heck, it wasn’t even complete for Office 97.

    I have nothing on the NSF format.  But if anyone has genuine interest, then send me an email.  I can hook you up with the relevant product managers covering Notes/Domino programmability.

  6. Alex says:

    Rob, "programmability" means "pay up for the right to write software to convert your documents". MSO documents are accessible in this sense since… well, ever.

  7. Rob Weir says:

    Actually "programmability" is a generic term we use for all interfaces that allow 3rd party developers to interact with IBM technology.  This could include standards, format documentation,  API’s, toolkits, sample code, etc. "Programmability" is an umbrella term we use for that whole range of interfaces.  

    What does Microsoft call it?  Maybe Developer Relations?  Same thing, different name.

  8. Dave S. says:

    "Note also that Microsoft’s claim for OOXML is that it is valuable because it is 100% compatible with the legacy formats."

    It’s not 100% compatible because the legacy applications cannot open files in the MSO-XML format.

    Some versions of the legacy apps can have newly created adapters added, but then they are new, hybrid apps.

  9. Dave,

    Given all the fuss in the previous months, I thought that we somewhere in those months had already hit an all-time low with regards to quality in the debate. I am kindda sad that you, with your comment, proved me wrong.


  10. Ian Easson says:

    Dave S:

    Your comment is (1) Irrelevant, and (2) Wrong.

    It is irrelevant because the quote you gave is about comparing a file format to a file format.  Whether or not any given application can open up either of those formats does not make a hill of beans to an argument about 100% compatability of formats!

    It is wrong for two reasons:

    (a) The "some" version of legacy apps you refer to is for all version of Microsoft Office from 97 and above (so you are guilty of trying to minimize things), and

    (b) The "new, hybrid apps" characterization is wrong.  Adding a small add-on to allow reading and writing in a different file format (e.g., Wordperfect or OOXML) does not make a release of Microsoft Office into a new application.  There are literally dozens of such small file format add-ons available, most from Microsoft, and they do not change the behaviour of the application other than for their intended purpose.