web metrics
The value of a checkbox as defined by a standard - Gray Matter - Site Home - TechNet Blogs

Gray Matter

Gray Knowlton's blog on Microsoft Office

The value of a checkbox as defined by a standard

The value of a checkbox as defined by a standard

  • Comments 4
  • Likes

Just keeping tabs on the state of matters for document format standards. I thought I'd share some of those interesting data points and connect some dots along the way.

Ouch. Rich Jelliffe discusses some of the dynamics of the OASIS ODF TC.

Rick says: "It seems that OASIS rules actually ban Technical Committee members from participating on the comments list with non-committee members. Communication is a one-way affair, an offering to silent gods. …

It didn't used to be this way. And snooping around the archives, it seems that when the rules were changed (by the then OASIS board) in 2003, there was quite a stink. The justification was to prevent spam: I cannot quite understand the logic, but the effect is that it meant if you want to participate at OASIS you have to pay."

I don't pretend to understand the inner workings of standards organizations, but I definitely recall some very sharp criticism of Microsoft and ECMA for some very similar reasons. Does this mean the withdraw of IBM from OASIS is imminent? J It's also strange how, when faced with criticism, some individuals revert to Microsoft-bashing as a means to defending their own actions. So then let me revert to some old habits and point out an apparent contrast between the words written in a different forum:

"Those who control the exchange format, can control interoperability and turn it on or off like a water faucet to meet their business objectives."this from the chair of the ODF TC.

Interesting: Alex Brown (a while ago, I know) drills on conformance issues for ODF 1.2 and exposes a fairly basic and problematic reality. Rick had a little to say about this as well.

This is an extension of the discussion that is exemplified by Oracle's comment on Open XML conformance during the BRM process, and those of Google as well. It seems that after all this time we're not settled on basic questions, for example whether or not all content within the ZIP package should be described using XML. Again I don't pretend to have an answer, but appreciate the level of attention that Open XML has brought to this problem.

The University of Illinois Law has published a paper on the (recent, but somewhat aged) state of interoperability for Open XML and ODF. It isn't a glowingly positive report for either standard, but it is based on older implementations, and by virtue of the fact that the numbers in the tables seem higher for Open XML than they do ODF, one could regard that as a "win" for Open XML. I think the bigger message is that for both Open XML and ODF, things are improving. In the long view, this report isn't surprising. In the 20+ year history of desktop business productivity applications, it is reasonable to expect a long ramp for interoperability for brand new document formats. I would like to raise my hand on this with a few questions:

Why isn't anyone talking about PDF? – seems like if one wanted to discuss document exchange, PDF would be an interesting basis for comparison. For example, how likely are various PDF Viewing applications to get it "right" when opening a PDF file? What is the degree of variance across the thousands of PDF generation implementations in the world? Odd that people seem to have vacated that space very quickly. The warm fire around which folks are huddled seems to be the one where people have a hope of editing.

Why isn't anyone talking about binary documents? – I hate to draw attention to a white elephant, but the use of binary document formats (at least by count of what is indexed on the web) is on the order of ~200,000 to 200,000,000. I would love to see an amendment to this study which illustrates the interoperability of binary documents in the same test.

And for all this, here's the point:

The rough edges of document format interoperability are apparent. One should question the fastest way to get to the goal of high quality inter application document exchange. It would seem that (reference above) that a wealth of IQ is being devoted to evolving standards to achieve the lofty goals that are floating around this topic; lots and lots of paint is being used for boxing things into tight corners.

For the folks who don't appreciate the handfighting and procedural aspects to the standards work, though, how can they get any benefit out of this? If it is the case that "the little guy" isn't welcome at the table in the standards discussion (which I don't believe is the case for Open XML or ODF), what is the fastest means to high-quality document exchange?

As far as I can see, only one vendor is doing this: http://www.documentinteropinitiative.org. I have not found this documentation from other developers: http://www.documentinteropinitiative.org/ECMA-376/reference.aspx, http://www.documentinteropinitiative.org/OASISODF1.1/reference.aspx. I am hopeful to see other vendors bringing this data to the table, beyond their participation in DII and on OpenXMLDeveloper.org. I hope to see this because it helps folks sort through the real problems vs. those that are less significant, and helps ensure that we're really talking about document format exchange, rather than hopeful RFP checkbox filling.

I am optimistic for the future and hopeful for more developer-to-developer engagement on real product implementations. I think it matters a great deal for this transition to standardized document formats.

Comments
Your comment has been posted.   Close
Thank you, your comment requires moderation so it may take a while to appear.   Close
Leave a Comment