Friday, February 01, 2008

Brian Jones: Open XML Formats : Harmonization: Finding the Differences

A timely reality check from Brian Jones; see the full post for more context-setting

There is a lot of interest in the harmonization topic, but at this point, it's not clear what the suggestion actually entails as Rob Weir himself acknowledged. Some people (including Rob) have suggested there is 90% overlap between the two standards and that harmonization should be pretty simple additive process where features of one spec are added to the other spec. I don't personally believe this; I've posted before about some pretty big differences in approach (in formulas and spreadsheets performance for example). Rationalizing these is not up to me or Rob Weir (or any specific company). The formats are not sub-sets and super-sets of each other, they are fundamentally different. Any effort for harmonization must begin with some deep thinking about how things like text, tables, styles, graphics and page layout models are different (finding the difference is at the core of the DIN work) and how they can be rationalized when their core design is very different.

My $.02: "harmonization" would essentially be a political appeasement move -- fine as long as 1) it doesn't unacceptably complicate the resulting model (although I'm far from certain that's possible...) and 2) it gets people to stop arguing about the relative merits of two very different alternatives, and to focus more on real-world customer requirements.

Brian Jones: Open XML Formats : Harmonization: Finding the Differences

No comments: